{"id":"https://openalex.org/W4409045922","doi":"https://doi.org/10.1145/3725843.3756041","title":"Dissecting and Modeling the Architecture of Modern GPU Cores","display_name":"Dissecting and Modeling the Architecture of Modern GPU Cores","publication_year":2025,"publication_date":"2025-10-17","ids":{"openalex":"https://openalex.org/W4409045922","doi":"https://doi.org/10.1145/3725843.3756041"},"language":"en","primary_location":{"id":"doi:10.1145/3725843.3756041","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3725843.3756041","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2503.20481","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102982627","display_name":"Rodrigo Huerta","orcid":"https://orcid.org/0000-0003-0052-7710"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Rodrigo Huerta","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"raw_orcid":"https://orcid.org/0000-0003-0052-7710","affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092217297","display_name":"Mojtaba Abaie Shoushtary","orcid":"https://orcid.org/0000-0003-2377-6939"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Mojtaba Abaie Shoushtary","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"raw_orcid":"https://orcid.org/0000-0003-2377-6939","affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114007075","display_name":"Jos\u00e9-Lorenzo Cruz","orcid":null},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Jos\u00e9-Lorenzo Cruz","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"raw_orcid":"https://orcid.org/0000-0001-5325-9153","affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100733331","display_name":"Antonio Gonz\u00e1lez","orcid":"https://orcid.org/0000-0002-0009-0996"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Antonio Gonzalez","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain"],"raw_orcid":"https://orcid.org/0000-0002-0009-0996","affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Catalunya, Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5102982627"],"corresponding_institution_ids":["https://openalex.org/I9617848"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.06139106,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"369","last_page":"384"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10741","display_name":"Video Coding and Compression Technologies","score":0.5552999973297119,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10741","display_name":"Video Coding and Compression Technologies","score":0.5552999973297119,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.5365999937057495,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5705390572547913},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.503335177898407},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4821029007434845},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3949144780635834}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5705390572547913},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.503335177898407},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4821029007434845},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3949144780635834}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3725843.3756041","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3725843.3756041","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2503.20481","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.20481","pdf_url":"https://arxiv.org/pdf/2503.20481","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null},{"id":"doi:10.48550/arxiv.2503.20481","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.20481","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2503.20481","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.20481","pdf_url":"https://arxiv.org/pdf/2503.20481","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":54,"referenced_works":["https://openalex.org/W4401211856","https://openalex.org/W4308090436","https://openalex.org/W3206534076","https://openalex.org/W2946072469","https://openalex.org/W2786171709","https://openalex.org/W1979527452","https://openalex.org/W4360831990","https://openalex.org/W2147657366","https://openalex.org/W3004853346","https://openalex.org/W1997162567","https://openalex.org/W1997523440","https://openalex.org/W3100384985","https://openalex.org/W2021211271","https://openalex.org/W2080592089","https://openalex.org/W2798724095","https://openalex.org/W3100327344","https://openalex.org/W3021029305","https://openalex.org/W2096661534","https://openalex.org/W2022632182","https://openalex.org/W2128120785","https://openalex.org/W2787181861","https://openalex.org/W4407848455","https://openalex.org/W2794501504","https://openalex.org/W1954144304","https://openalex.org/W4404954401","https://openalex.org/W4247641071","https://openalex.org/W2963240192","https://openalex.org/W3102510044","https://openalex.org/W2345041994","https://openalex.org/W2155503253","https://openalex.org/W2000636639","https://openalex.org/W4409133360","https://openalex.org/W4400409880","https://openalex.org/W2791673912","https://openalex.org/W2907701003","https://openalex.org/W2296602564","https://openalex.org/W2466930524","https://openalex.org/W2963989532","https://openalex.org/W2132258839","https://openalex.org/W2209148515","https://openalex.org/W2047060659","https://openalex.org/W2789977517","https://openalex.org/W2620382510","https://openalex.org/W4309672181","https://openalex.org/W2952928793","https://openalex.org/W3157055696","https://openalex.org/W2979340153","https://openalex.org/W2029940394","https://openalex.org/W2901206390","https://openalex.org/W2110195531","https://openalex.org/W2163396412","https://openalex.org/W3043303806","https://openalex.org/W2580538010","https://openalex.org/W4388858680"],"related_works":["https://openalex.org/W2899084033","https://openalex.org/W2005148983","https://openalex.org/W2012954338","https://openalex.org/W2096672917","https://openalex.org/W2392023973","https://openalex.org/W3189307731","https://openalex.org/W1428699136","https://openalex.org/W2949962288","https://openalex.org/W2364686214","https://openalex.org/W1998560227"],"abstract_inverted_index":{"GPUs":[0,57,119,217],"are":[1,33],"the":[2,62,73,78,81,84,87,98,129,132,137,208],"most":[3,18],"popular":[4],"platform":[5],"for":[6,197],"accelerating":[7],"HPC":[8],"workloads,":[9],"such":[10,201],"as":[11,202],"artificial":[12],"intelligence":[13],"and":[14,54,90,94,120,136,148,229],"science":[15],"simulations.":[16],"However,":[17],"microarchitectural":[19,156],"research":[20],"in":[21,167,175,214,225],"academia":[22],"relies":[23],"on":[24,30,110,144,223],"GPU":[25,45],"core":[26],"pipeline":[27],"designs":[28],"based":[29,109,222],"architectures":[31],"that":[32,192,207],"more":[34],"than":[35,170],"15":[36],"years":[37],"old.":[38],"This":[39],"paper":[40],"reverse":[41],"engineers":[42],"modern":[43,117,215],"NVIDIA":[44,118,199,216],"cores,":[46],"unveiling":[47],"many":[48],"key":[49],"aspects":[50],"of":[51,80,86,97,131,139,178,227],"its":[52,91],"design":[53],"explaining":[55],"how":[56,72,104],"leverage":[58],"hardware-compiler":[59],"techniques":[60],"where":[61],"compiler":[63],"guides":[64],"hardware":[65,185,220],"during":[66],"execution.":[67],"In":[68],"particular,":[69],"it":[70,102],"reveals":[71],"issue":[74,82],"logic":[75],"works":[76],"including":[77],"policy":[79],"scheduler,":[83],"structure":[85],"register":[88,133,140],"file":[89,134,141],"associated":[92],"cache,":[93],"multiple":[95],"features":[96],"memory":[99],"pipeline.":[100],"Moreover,":[101],"analyses":[103],"a":[105,111,219],"simple":[106],"instruction":[107],"prefetcher":[108],"stream":[112],"buffer":[113],"fits":[114],"well":[115],"with":[116,181],"is":[121],"likely":[122],"to":[123,183],"be":[124],"used.":[125],"Furthermore,":[126],"we":[127,158,190,205],"investigate":[128],"impact":[130],"cache":[135],"number":[138],"read":[142],"ports":[143],"both":[145],"simulation":[146],"accuracy":[147],"performance.":[149],"By":[150],"modeling":[151],"all":[152],"these":[153],"new":[154,194],"discovered":[155],"details,":[157],"achieve":[159],"18.24%":[160],"lower":[161],"mean":[162],"absolute":[163],"percentage":[164],"error":[165],"(MAPE)":[166],"execution":[168],"cycles":[169],"previous":[171],"state-of-the-art":[172],"simulators,":[173],"resulting":[174],"an":[176],"average":[177],"13.98%":[179],"MAPE":[180],"respect":[182],"real":[184],"(NVIDIA":[186],"RTX":[187],"A6000).":[188],"Also,":[189],"demonstrate":[191],"this":[193],"model":[195],"stands":[196],"other":[198],"architectures,":[200],"Turing.":[203],"Finally,":[204],"show":[206],"software-based":[209],"dependence":[210],"management":[211],"mechanism":[212,221],"included":[213],"outperforms":[218],"scoreboards":[224],"terms":[226],"performance":[228],"area.":[230]},"counts_by_year":[],"updated_date":"2026-05-21T09:19:25.381259","created_date":"2025-04-01T00:00:00"}